|
Vol.7, No.2, May 2018. ISSN: 2217-8309 eISSN: 2217-8333
TEM Journal
TECHNOLOGY, EDUCATION, MANAGEMENT, INFORMATICS Association for Information Communication Technology Education and Science |
Majority Vote of Ensemble Machine Learning Methods for Real-Time Epilepsy Prediction Applied on EEG Pediatric Data
Samed Jukić, Dino Кеčo, Jasmin Kevrić
© 2018 Samed Jukić, published by UIKTEN. This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. (CC BY-NC-ND 4.0)
Citation Information: TEM Journal. Volume 7, Issue 2, Pages 313-318, ISSN 2217-8309, DOI: 10.18421/TEM72-11, May 2018.
Received: 09 March 2018
Abstract:
The main aim of the study is to develop a real-time epilepsy prediction approach by using the ensemble machine learning techniques that might predict offline seizure paradigms. The proposed seizure prediction algorithm is patient-specific since generalization showed no satisfactory results in our previous studies. The algorithm is tested on CHB-MIT database comprised of EEG data from pediatric epileptic patients. Based on relations to number of seizures and number of files, gender and age, three patients have been chosen for this study. The special majority voting algorithm is proposed and used for raising an alarm of upcoming seizure. EEG signals are denoised using MSPCA (Multiscale PCA), the features were extracted by WPD (wavelet packet decomposition), and EEG signals were classified using Rotation Forest. The significance of the study lies in the fact that the proposed seizure prediction algorithm could be used in novel diagnostic and therapeutic applications for pediatric patients.
Keywords –Majority Vote, Rotation Forest, Real-Time Prediction, Epilepsy. |
|
----------------------------------------------------------------------------------------------------------- ----------------------------------------------------------------------------------------------------------- |